The overall goal of our research is to contribute to understanding the role of the kidneys in the maintenance of body fluid volume and arterial blood pressure. Specifically, we study the renal mechanisms responsible for the match between glomerular filtration of plasma and tubular salt and fluid reabsorption that is ultimately responsible for the precise balance between salt intake and excretion. One of the key mechanisms that mediates between renal absorption and filtration function is the so-called tubuloglomerular feedback, a regulatory pathway in which a measure of tubular reabsorptive capacity serves as the signal to alter glomerular filtration. 1. An attempt was made to improve understanding of the renovascular actions of adenosine as the agent responsible for tubuloglomerular feedback (TGF)-induced vasoconstriction. Studies of the mechanism of adenosine-induced vasoconstriction in isolated perfused afferent arterioles indicate that the constrictor response to adenosine is initiated by adenosine 1 receptors (A1AR) coupling to a PTX sensitive Galphai-protein, and subsequent activation of PLC, presumably through ?O???nsubunits released from Galphai. Ca influx is maintained by activation of voltage dependent Ca channels following depolarization caused most likely through Ca-activated Cl channels. Despite its vasoconstrictor function the steady-state effect of intravenous adenosine is renal vasodilatation, an effect that is augmented in A1AR-/- mice. However, when adenosine was delivered to the interstitium under a laser Doppler probe measuring superficial capillary blood flow adenosine or CHA caused vasoconstriction. In isolated perfused afferent arterioles of A1AR+/+, addition of adenosine to the bath at 10-7 M constricted the vessel while addition of adenosine to the perfusate had no significant effect. The observed sidedness suggests a role of the endothelium in the vasodilator action of adenosine. Since isolated blood vessels constrict to intravascular adenosine in the presence of the NOS inhibitor L-NAME, it appears that adenosine causes the production of nitric oxide, and that this contributes to its vasodilator action. Thus, the vascular effects of adenosine are to some extent endothelium-dependent. 2. To further study the role of the TGF in the maintenance of body salt balance, we have generated A1AR-/-mice on an aquaporin 1 knockout background. These A1AR/AQP1 double knockout mice have a marked reduction of proximal tubular fluid reabsorption, and because of the absence of TGF a doubling in salt delivery to the distal nephron. However, despite this distal fluid overload they are not overtly salt-losing indicating that NaCl absorption along the distal convoluted tubule and collecting duct can substitute for TGF-induced reductions in tubular salt load. 3. Adenosine mediating the TGF response may be either generated by macula densa cells or by extracellular degradation of released ATP through the action of 5!| nucleotidase. To distinguish between these possibilities we have generated 5!| nucleotidase knockout mice. We have obtained germline transmission of the knockout mutation and are in the process of crossing heterozygous offspring. 4. Renin formation in granular cells of the juxtaglomerular apparatus plays a critical role in the maintenance of body fluid volume and arterial blood pressure. To be able to delete specific gene products from granular cells we are working on directing the expression of Cre recombinase to these cells. We have developed mouse lines expressing Cre recombinase mRNA in a tissue-specific way using a 4.2 kb fragment of the murine renin promoter, but have been unable to confirm the presence of tissue specific Cre activity by crossing these lines into the ROSA26-GFP reporter strain background. We now have made a vector in which Cre is driven by a 12 kb human renin promoter that has previously been shown to direct LacZ expression to granular cells. The generation of transgenic mice is under way. 5. The activity of the Na,K,2Cl cotransporter NKCC2 is critical for the initiation of the signaling events by which the macula densa cells affect renin secretion and vascular tone. NKCC2 knockout mice have been made, but these animals are not useful for physiologic studies because of massive early postnatal fluid losses and death. Since NKCC2 undergoes alternative splicing into three isoforms (A, B, and F), and since there is cell-specific expression of different isoforms along the tubule, we are currently in the process of making NKCC2 isoform-specific knockouts. We have generated three constructs, each consisting of the following components: about 5 kb of 5' homologous sequence-Flag peptide-stop codon-floxed NeoR-about 5 kb of 3' homologous sequence. The insertion of these constructs is either in the A, B, or F region of exon 4. After stem cell transfection we have been able to identify 2 positive clones for the A and F mutation, and 3 clones for the B mutation.